VLA Foundry: A Unified Framework for Training Vision-Language-Action Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918460068462592 |
|---|---|
| author | Mercat, Jean Keh, Sedrick Arora, Kushal Huang, Isabella Shah, Paarth Nishimura, Haruki Iwase, Shun Liu, Katherine |
| author_facet | Mercat, Jean Keh, Sedrick Arora, Kushal Huang, Isabella Shah, Paarth Nishimura, Haruki Iwase, Shun Liu, Katherine |
| contents | We present VLA Foundry, an open-source framework that unifies LLM, VLM, and VLA training in a single codebase. Most open-source VLA efforts specialize on the action training stage, often stitching together incompatible pretraining pipelines. VLA Foundry instead provides a shared training stack with end-to-end control, from language pretraining to action-expert fine-tuning. VLA Foundry supports both from-scratch training and pretrained backbones from Hugging Face. To demonstrate the utility of our framework, we train and release two types of models: the first trained fully from scratch through our LLM-->VLM-->VLA pipeline and the second built on the pretrained Qwen3-VL backbone. We evaluate closed-loop policy performance of both models on LBM Eval, an open-data, open-source simulator. We also contribute usability improvements to the simulator and the STEP analysis tools for easier public use. In the nominal evaluation setting, our fully-open from-scratch model is on par with our prior closed-source work and substituting in the Qwen3-VL backbone leads to a strong multi-task table top manipulation policy outperforming our baseline by a wide margin. The VLA Foundry codebase is available at https://github.com/TRI-ML/vla_foundry and all multi-task model weights are released on https://huggingface.co/collections/TRI-ML/vla-foundry. Additional qualitative videos are available on the project website https://tri-ml.github.io/vla_foundry. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_19728 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | VLA Foundry: A Unified Framework for Training Vision-Language-Action Models Mercat, Jean Keh, Sedrick Arora, Kushal Huang, Isabella Shah, Paarth Nishimura, Haruki Iwase, Shun Liu, Katherine Robotics Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning Software Engineering I.2.9; I.2.6; I.2.7; I.2.10 We present VLA Foundry, an open-source framework that unifies LLM, VLM, and VLA training in a single codebase. Most open-source VLA efforts specialize on the action training stage, often stitching together incompatible pretraining pipelines. VLA Foundry instead provides a shared training stack with end-to-end control, from language pretraining to action-expert fine-tuning. VLA Foundry supports both from-scratch training and pretrained backbones from Hugging Face. To demonstrate the utility of our framework, we train and release two types of models: the first trained fully from scratch through our LLM-->VLM-->VLA pipeline and the second built on the pretrained Qwen3-VL backbone. We evaluate closed-loop policy performance of both models on LBM Eval, an open-data, open-source simulator. We also contribute usability improvements to the simulator and the STEP analysis tools for easier public use. In the nominal evaluation setting, our fully-open from-scratch model is on par with our prior closed-source work and substituting in the Qwen3-VL backbone leads to a strong multi-task table top manipulation policy outperforming our baseline by a wide margin. The VLA Foundry codebase is available at https://github.com/TRI-ML/vla_foundry and all multi-task model weights are released on https://huggingface.co/collections/TRI-ML/vla-foundry. Additional qualitative videos are available on the project website https://tri-ml.github.io/vla_foundry. |
| title | VLA Foundry: A Unified Framework for Training Vision-Language-Action Models |
| topic | Robotics Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning Software Engineering I.2.9; I.2.6; I.2.7; I.2.10 |
| url | https://arxiv.org/abs/2604.19728 |